Multivalued Ferroelectric Associative Memory Design
نویسنده
چکیده
This talk provides an introduction to Content Addressable Memories (CAM), Ferroelectric memories, and Multiple-Valued Ferroelectric Content-Addressable Memories (MVFCAM). Circuit simulation results are presented for two architectures of MVFCAM. What is an associative memory? An associative memory, or a Content Addressable Memory (CAM), searches for data by content rather than by address [1-4]. Thus, a CAM is different from a conventional address-based RAM which searches for data at a particular address. One application of CAMs is in an electronic spelling checker, where the CAM is searched to locate a word with a specific spelling. If there is at least one word in the CAM with a matched spelling, the search is successful. A search by content can be performed in a word-serial or a bit-serial approach. In a word-serial approach, the words in CAM are compared sequentially against the input word for a possible match. Assuming a CAM with w words, a word-serial approach requires w comparison steps to complete a search. In a bit-serial approach, the most significant bit of each word in CAM is compared simultaneously against the most significant bit of the input word. In the next steps, bits of less significance are compared sequentially against the corresponding bits of the input word. Assuming a CAM with n-bit words, a bit-serial search approach requires n comparison steps to complete a search. As a result, in a CAM with w n-bit words, a bit-serial approach is advantageous in terms of the search speed if n << w. Why ‘multilevel’ CAM? In a bit-serial approach, an m-level data storage can ideally increase both the search speed and the CAM density by a factor of (log2m). The speed or the density is usually compromised to some extent for a less complex design. Multivalued Ferroelectric Associative Memory Design Ali Sheikholeslami and P. Glenn Gulak Department of Electrical and Computer Engineering, University of Toronto, Canada May 27, 1997 7th International Workshop on Post-Binary ULSI Systems, Fukuoka, Japan, May 26, 1998 69 What is a ferroelectric capacitor? A ferroelectric capacitor is physically distinguished from a regular capacitor by substituting the dielectric with a ferroelectric material [5-6]. A ferroelectric capacitor is capable of storing one bit of data in the form of electric polarization. A positive or negative voltage pulse brings the capacitor to its positive or negative polarization state, respectively. The capacitor retains its polarization state without a power supply. A ferroelectric capacitor constitutes the core of a storage cell in a Ferroelectric Random Access Memory (FRAM). What is a multilevel FCAM? Ferroelectric capacitors are used as the storage elements in the recently proposed CAM architectures [2] to implement a Multiple-Valued Ferroelectric CAM (MVFCAM). A 4-level parallel-access FCAM, for example, is capable of storing 4-level data, and searching for data that is less than, greater than, or equal to the input data. Moreover, the search is twice as fast as a search in a binary FCAM. The proposed architectures are also write programmable in real time; a feature that is absent in most nonvolatile CAMs.
منابع مشابه
Asymptotically Stable Multi-valued Many-to-many Associative Memory Neural Network and Its Application in Image Retrieval
As an important artificial neural network, associative memory model can be employed to mimic human thinking and machine intelligence. In this paper, first, a multi-valued many-to-many Gaussian associative memory model (MGAM) is proposed by introducing the Gaussian unidirectional associative memory model (GUAM) and Gaussian bidirectional associative memory model (GBAM) into Hattori et al’s multi...
متن کاملAssociative learning and memory duration of Trichogramma brassicae
Learning ability and memory duration are two inseparable factors which can increase theefficiency of a living organism during its lifetime. Trichgramma brassice Bezdenko (Hym.:Trichogrammatidae) is a biological control agent widely used against different pest species.This research was conducted to study the olfactory associative learning ability and memoryduration of T. brassicae under laborato...
متن کاملOn the Recall Capability of Recurrent Exponential Fuzzy Associative Memories Based on Similarity Measures
Recurrent exponential fuzzy associative memories (RE-FAMs) are non-distributive memory models derived from the multivalued exponential recurrent associative memory (MERAM) of Chiueh and Tsai. A RE-FAM defines recursively a sequence of fuzzy sets obtained by a weighted average of the fundamental memories. In this paper, we show that the output of a single-step RE-FAM can be made as close as desi...
متن کاملBottom-up Approach for Generating Regular Patterns of Nanoscaled Ferroelectric Perovskites
Ferroelectric materials are of growing interest in international research and development. Special attention is drawn to the integration of ferroelectric materials in non-volatile random access memory cells (FeRAMs). For future high-density FeRAMs, the scaling of the size of ferroelectric structures needs to be understood in order to elucidate possible limitations to memory densities. Various e...
متن کاملA Morphological Associative Memory Employing A Stored Pattern Independent Kernel Image and Its Hardware Model
An associative memory provides a convenient way for pattern retrieval and restoration, which has an important role for handling data distorted with noise. As an effective associative memory, we paid attention to a morphological associative memory (MAM) proposed by Ritter. The model is superior to ordinary associative memory models in terms of calculation amount, memory capacity, and perfect rec...
متن کامل